Date post: | 07-Jul-2018 |
Category: |
Documents |
Upload: | mose-delpozzo |
View: | 261 times |
Download: | 0 times |
of 74
8/19/2019 Classical Mechanics Lectures
1/74
Lecture Notes
PHY 321 - Classical Mechanics I www.pa.msu.edu/courses/phy321
Instructor: Scott Pratt, [email protected]
These notes are NOT meant to be a substitute for the book. The text Classical Mechanics by JohnR. Taylor is required. The course will follow the order of material in the text. These notes areonly meant to outline what was covered in class and to provide something students can printand annotate rather than taking full lecture notes.
Contents
1 Math Basics 1
1.1 Scalars, Vectors and Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Newton’s Laws 8
2.1 Air Resistance in One Dimension . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2 Projectile Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Motion in a Magnetic Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.4 Conservation of Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.5 Solving an integral numerically . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.6 Conservation of Momentum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.7 Conservation of Angular Momentum . . . . . . . . . . . . . . . . . . . . . . . . 18
2.8 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3 Oscillations 23
3.1 Harmonic Oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.2 Damped Oscillators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.3 Sinusoidally Driven Oscillators . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.4 Alternative Derivation for Driven Oscillators . . . . . . . . . . . . . . . . . . . . 27
3.5 Resonance Widths; the Q factor . . . . . . . . . . . . . . . . . . . . . . . . . . . 293.6 Principal of Superposition and Periodic Forces . . . . . . . . . . . . . . . . . . . 30
3.7 Response to arbitrary, but temporary, force . . . . . . . . . . . . . . . . . . . . . 32
3.8 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4 Gravity and Central Forces 36
4.1 Gravity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
http://www.pa.msu.edu/courses/phy321http://www.amazon.com/Classical-Mechanics-John-R-Taylor/dp/189138922Xhttp://www.amazon.com/Classical-Mechanics-John-R-Taylor/dp/189138922Xhttp://www.pa.msu.edu/courses/phy321
8/19/2019 Classical Mechanics Lectures
2/74
4.2 Tidal Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.3 Deriving elliptical orbits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.4 Effective or Centrifugal Potential . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.5 Stability of Orbits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.6 Scattering and Cross Sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.7 Center-of-mass Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464.8 Rutherford Scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.9 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5 Rotating Coordinate Systems 53
5.1 Accelerating Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.2 Rotating Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.3 Coriolis Force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5.4 The Foucault Pendulum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
6 Lagrangians 58
6.1 Calculus of Variations and the Euler Equation . . . . . . . . . . . . . . . . . . . 58
6.2 Adding constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
6.3 Lagrangians . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
6.4 Proving Lagrange’s equations of motion from Newton’s Laws . . . . . . . . . . . 63
6.5 Conservation Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
6.6 Numerically Solving Differential Equations . . . . . . . . . . . . . . . . . . . . . 696.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
8/19/2019 Classical Mechanics Lectures
3/74
PHY 321 Lecture Notes 1 MATH BASICS
1 Math Basics
1.1 Scalars, Vectors and Matrices
A scalar is something with a value that is independent of coordinate system. Examples are mass,
or the relative time between events. A vector has magnitude and direction. Under rotation, themagnitude stays the same but the direction changes. Scalars have no spatial index, whereas athree-dimensional vector has 3 indices, e.g. the position r has components r1, r2, r3, which areoften referred to as x ,y ,z.
There are several categories of changes of coordinate system. The observer can translate theorigin, the observer might suddenly be moving with constant velocity, or the observer mightrotate his coordinate axes. For instance, a particle’s position vector changes when the origin istranslated, but its velocity does not. When you study relativity you will find that quantities youthought of as scalars, such as time or electric potential, are actually parts of four-dimensionalvectors and that changes of the velocity of the reference frame act in a similar way to rotations.
In addition to vectors and scalars, there are matrices, which have two indices. One also hasobjects with 3 or four indices. These are called tensors of rank n, where n is the number of indices. A matrix is a rank-two tensor. The Levi-Civita symbol, ijk used for cross products, is athird-rank tensor.
Unit Vectors
Also known as basis vectors, unit vectors point in the direction of the coordinate axes, have unitnorm, and are orthogonal to one another. Sometimes this is referred to as an orthonormal basis,
êi · ê j = δij = 1 0 0
0 1 00 0 1
. (1.1)
Here, δij is unity when i = j and is zero otherwise. This is called the unit matrix, becauseyou can multiply it with any other matrix and not change the matrix. The "dot" denotes the dot
product, A · B = A1B1 + A2B2 + A3B3 = |A||B| cos θAB. Sometimes the unit vectors arecalled x̂, ŷ and ẑ. Vectors can be decomposed in terms of unit vectors,
r = r1ê1 + r2ê2 + r3ê3. (1.2)
The vector components r1, r2 and r3 might be called x, y and z for a displacement of vx, vy andv
z for a velocity.
Rotations
We use rotations as an example of matrices and their operations. One can consider a differentorthonormal basis ê1, ê
2 and ê
3. The same vector r mentioned above can also be expressed in
the new basis,r = r 1ê
1 + r
2ê2 + r
3ê3. (1.3)
1
8/19/2019 Classical Mechanics Lectures
4/74
PHY 321 Lecture Notes 1 MATH BASICS
Even though it is the same vector, the components have changed. Each new unit vector êi can be expressed as a linear sum of the previous vectors,
êi = j
λij ê j, (1.4)
and the matrix λ can be found by taking the dot product of both sides with êk,
êk · êi = j
λijêk · ê j
êk · êi = j
λijδ jk = λik. (1.5)
Thus, the matrix lambda has components λij that are equal to the cosine of the angle betweennew unit vector êi and the old unit vector ê j .
λ = ê
1 · ê1 ê1 · ê2 ê1 · ê3
ê
2 · ê1 ê
2 · ê2 ê
2 · ê3ê3 · ê1 ê3 · ê2 ê3 · ê3
, λij = cos θij. (1.6)
Note that the matrix is not symmetric, λij = λ ji .One can now return to the equivalence of the two vectors to find how the new components andthe old components are related.
r1ê1 + r2ê2 + r3ê3 = r1ê1 + r
2ê2 + r
3ê3 (1.7)
= j
r1λ1 jê j + r2λ2 jê j + r
3λ3 jê j.
For this equivalence to be correct the prefactor of each unit vector must be equivalent on bothsides, so
ri = j
λ jir j =
j
λtijr j. (1.8)
Here transpose is denoted by the subscript t and involves simply switching the indices of amatrix, M tij ≡ M ji . One could consider the inverse procedure, where one switched the primedand unprimed quantities. The matrix λ would then correspond to the inverse transformation,
ri = j
λt,−1ij r j. (1.9)
In this case the components are λ−1ij = êi · ê j as opposed to what we found above, λij = êiê j .By inspection, one can then see that
λ−1 = λt. (1.10)
This means thatri =
j
λijr j. (1.11)
2
8/19/2019 Classical Mechanics Lectures
5/74
PHY 321 Lecture Notes 1 MATH BASICS
When the inverse of a real square matrix equals its transpose, one would call λ a unitary or or-thogonal matrix. Note that rotating a vector by an angle φ is equivalent to rotating the coordinatesystem by the opposite angle.
Example 1.1:Find the rotation matrix λ for finding the components in the primed coordinate system given
from those in the unprimed system, given that the unit vectors in the new system are found byrotating the coordinate system by and angle φ about the z axis.Solution:In this case
ê1 = cos φê1 − sin φê2,ê2 = sin φê1 + cos φê2,
ê3 = ê3.
By inspecting Eq. (1.5),
λ = cos φ − sin φ 0sin φ cos φ 0
0 0 1
.
Vector and Matrix Operations
• Scalar Product (or dot product): For vectors A and B,
A·
B = i
AiB
i =
|A
||B
|cos θ
AB,
|A| ≡
A · A.
• Multiplying a matrix C times a vector A:
(CA)i = j
C ijA j.
For the ith element one takes the scalar product of the ith row of C with the vector A.
• Mutiplying matrices C and D:
(CD)ij =
k
C ikDkj,
This means the one obtains the ij element by taking the scalar product of the ith of C withthe jth column of D.
3
8/19/2019 Classical Mechanics Lectures
6/74
PHY 321 Lecture Notes 1 MATH BASICS
• Vector Product (or cross product) of vectors A and B:
C = A × B,C i = ijkA jBk.
Here is the third-rank anti-symmetric tensor, also known as the Levi-Civita symbol. Itis
±1 only if all three indices are different, and is zero otherwise. The choice of
±1 de-
pends on whether the indices are an even or odd permutation of the original symbols. Thepermutation xyz or 123 is considered to be +1. For the 27 elements,
123 = 231 = 312 = 1,
213 = 132 = 321 = −1,iij = iji = jii = 0.
You used cross products extensively when studying magnetic fields. Since the matrix isanti-symmetric, switching the x and y axes (or any two axes) flips the sign. If the coordi-nate system is right-handed, meaning the xyz axes satisfy x̂× ŷ = ẑ, where you can pointalong the x axis with your extended right index finger, the y axis with your contractedmiddle finger and the z axis with your extended thumb. Switching to a left-handed sys-
tem flips the sign of the vector C = A × B. Note that A × B = − B × A. The vector C is perpendicular to both A and B and the magnitude of C is given by
|C | = |A||B| sin θAB.Vectors obtained by the cross product of two real vectors are called pseudo-vectors becausethe assignment of their direction can be arbitrarily flipped by defining the Levi-Civita sym-
bol to be based on left-handed rules. Examples are the magnetic field and angular momen-tum. If the direction of a real vector prefers the right-handed over the left-handed direction,
that constitutes a violation of parity. For instance, one can polarize the spins (angular mo-mentum) of nuclei with a magnetic field so that the spins preferentially point along thedirection of the magnetic field. This does not violate parity since both are pseudo-vectors.Now assume these polarized nuclei decay and that electrons are one of the products. If these electrons prefer to exit the decay parallel vs. antiparallel to the polarizing magneticfield, this constitutes parity violation because the direction of the outgoing electron mo-menta are a real vector. This is exactly what is observed in weak decays.
• Differentiation of vector with respect to a scalar: For example, the acceleration is dv/dt:
(dv/dt)i = dvi
dt.
• Angular velocity: Choose the vector ω so that the motion is like your fingers wrappingaround your thumb on your right hand with ω pointing along your thumb. The magnitudeis dφ|/dt.
• Gradient operator ∇: This is the derivative ∂/∂x, ∂/∂y and ∂/∂z, where ∂ x means ∂/∂ x.For taking the gradient of a scalar Φ,
grad Φ, (∇Φ(x , y , z , t))i = ∂/∂riΦ(r, t).
4
8/19/2019 Classical Mechanics Lectures
7/74
PHY 321 Lecture Notes 1 MATH BASICS
For taking the dot product of the gradient with a vector, sometimes called a divergence,
div A, ∇ · A =
i
∂/∂riAi.
For taking the vector product with another vector, sometimes called curl, ∇ × A,
curl A, (∇ ×
A)i =
jk
ijk∂/∂r jAk(r, t).
• The Laplacian is referred to as ∇2 and is defined as
∇2 = ∇ · ∇ = ∂ 2
∂x2 +
∂ 2
∂y 2 +
∂ 2
∂z 2.
Some identities
Here we simply state these, but you may wish to prove a few. They are useful for this class and
will be essential when you study E&M. A · ( B × C ) = B · ( C × A) = C · ( A × B) (1.12)
A × ( B × C ) = ( A · C ) B − ( A · B) C ( A × B) · ( C × D) = ( A · C )( B · D) − ( A · D)( B · C )
Example 1.2:The height of a hill is given by the formula
z = 2xy − 3x2 − 4y2 − 18x + 28y + 12.Here z is the height in meters and x and y are the east-west and north-south coordinates. Findthe position x, y where the hill is the highest, and give its height.
Solution: The maxima or minima, or inflection points, are given when ∂ xz = 0 and ∂ yz = 0.
∂ xz = 2y − 6x − 18 = 0,∂ yz = 2x − 8y + 28 = 0.
Solving for two equations and two unknowns gives one solution x = −2, y = 3. This thengive z = 72. Although this procedure could have given a minimum or an inflection point youcan look at the form for z and see that for x = y = 0 the height is lower, therefore it is not a
minimum. You can also look and see that the quadratic contributions are negative so the heightfalls off to −∞ far away in any direction. Thus, there must be a maximum, and this must beit. Deciding between maximum or minimum or inflection point for a general problem wouldinvolve looking at the matrix ∂ i∂ jz at the specific point, then finding the eigenvalues of thematrix and seeing whether they were both positive (minimum) both negative (maximum) orone positive and one negative (saddle point).
5
8/19/2019 Classical Mechanics Lectures
8/74
PHY 321 Lecture Notes 1 MATH BASICS
Gauss’s Theorem and Stokes’s Theorem
For an integral over a volume V confined by a surface S , Gauss’s theorem gives V
dv∇ · A =
S
d S · A.
For a closed path C which carves out some area S , C
d · A =
S
ds · (∇ × A)
Stoke’s law can be understood by considering a small rectangle, −∆x < x < ∆x, −∆y <y
8/19/2019 Classical Mechanics Lectures
9/74
PHY 321 Lecture Notes 1 MATH BASICS
4. Multiply the rotation matrix in Example 1.1 by its transpose to show that the matrix isunitary, i.e. you get the unit matrix.
5. Find the matrix for rotating a coordinate system by 90 degrees about the x axis.
6. Consider a rotation where in the new coordinate system the three axes are at angles α, βand γ relative to the original x axis. Show that
cos2 α + cos2 β + cos2 γ = 1.
7. Find the rotation matrix for each of the three rotations, then find the matrix for the com- bined rotations. First rotate by 90◦ around the y axis to get to the primed system. Thenrotate the primed system 90◦ about the new z axis to get to the doubly primed system.Then find the matrix that rotates the doubly primed coordinate system 90◦ about the newx axis to get to the triply primed system. Finally, find the matrix that performs all threeoperations sequentially.
8. Consider a parity transformation which reflects about the x = 0 plane. Find the matrix that
performs the transformation. Find the matrix that performs the inverse transformation.
9. Show that the scalar product of two vectors is unchanged if both undergo the same rota-tion.
10. Show that k
ijkklm = δilδ jm − δimδ jl.
11. Consider a cubic volume V = L3 defined by 0 < x < L, 0 < y < L and 0 < z < L.
Consider a vector A that depends arbitrarily on x ,y ,z. Show how Gauss’s law, V
dv∇ · A =
S
d S · A,
is satisfied by direct integration.
12. Consider the function z = 3x2 − 4y2 + 12xy − 6x + 24. Find any maxima or minimaand determine whether it is a maximum or a minimum.
7
8/19/2019 Classical Mechanics Lectures
10/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
2 Newton’s Laws
In this chapter we mainly consider the motion of a single particle moving under the influenceof some set of forces. For Sec.s 2.1-2.3 we will consider some problems where the force does
not depend on the position. In that case Newton’s law ṁv = F (v) is a first-order differential
equation and one solves for v(t), then moves on to integrate v to get the position. In section 2.4we consider the case where the force depends on the position and we show how to use energyconservation to arrive at the answer, though as shown in Sec. 2.5, this may require numericalsolution. The role of momentum conservation in solving certain classes of problems is presentedin Sec. 2.6 and the role of angular momentum is shown in Sec. 2.7.
2.1 Air Resistance in One Dimension
Air resistance tends to scale as the square of the velocity. One can see this by considering themomentum imparted on the air molecules. If an object sweeps through a volume dV of air in
time dt, the momentum imparted on the air is
dP = ρmdV v, (2.1)
where v is the velocity of the object and ρm is the mass density of the air. If the molecules bounce back as opposed to stop you would double the size of the term. The opposite value of the momentum is imparted onto the object itself. Geometrically, the differential volume is
dV = Avdt, (2.2)
where A is the cross-sectional area and vdt is the distance the object moved in time dt. Plugging
this into the expression above, dP
dt= −ρmAv2. (2.3)
This is the force felt by the particle, and is opposite to its direction of motion. Now, since airdoesn’t stop when it hits an object, but flows around the best it can, the actual force is reduced
by a dimensionless factor cW , called the drag coefficient.
F drag = −cW ρmAv2, (2.4)
and the acceleration is
dv
dt = −cW ρmA
m v2. (2.5)
8
8/19/2019 Classical Mechanics Lectures
11/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
For a particle with initial velocity v0, one can separate the dt to one side of the equation, andmove everything with vs to the other side, then integrate,
dv
v2 = −Bdt, B ≡ cW ρmA
m. (2.6)
vf
v0
dv
v2
=
−Btf ,
1
v0− 1
vf
= −Btf , (2.7)
v = 1
1/v0 + Bt.
Here, I dropped the subscript f in the last expression.
Example 2.1:Consider a ball dropped from rest. Let B = cW ρmA/m represent the effect of the drag force as
above.
1. Solve for the terminal velocity.
Solution: This is where the acceleration is zero. The equations of motion are
dv
dt= −g + Bv2,
with the initial velocity set to zero. Since the motion is downward, the drag force is posi-tive. The terminal velocity is that for which the acceleration is zero,
vt =
g/B.
2. Find the velocity as a function of time.
Solution: Integrate as above,
−Bdt = dv(g/B) − v2 ,
−Bt = 1vt
v/vt0
dx 1
1 − x2
−Bt = 1
vt tanh−1
(v/vt),v = −vt tanh(Bvtt).
9
8/19/2019 Classical Mechanics Lectures
12/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
2.2 Projectile Motion
As an example of Newton’s Laws we consider projectile motion with a drag force. Even thoughair resistance is largely proportional to the square of the velocity, we will consider the drag force
to be linear to the velocity for the purposes of this exercise. The acceleration a = F /m then becomes
dvx
dt= −γvx, (2.8)
dvy
dt= −γvy − g.
Here, the drag force is γmv, so γ has dimensions of inverse time.
We will go over two different ways to solve this equation. The first by direct integration, andthe second as a differential equation. To do this by direct integration, one simply multiplies bothsides of the equations above by dt, then divide by the appropriate factors so that the vs are allon one side of the equation and the dt is on the other. For the x motion one finds an easilyintegrable equation,
dvx
vx= −γdt, (2.9) vfx
v0x
dvx
vx= −γ
tf 0
dt,
ln
vfx
v0x
= −γtf ,
vfx = v0xe−γt.
Here, I leave the subscript off the final time in the last expression. This is very much the result
you would have written down by inspection. For the y velocity,
dvy
vy + g/γ = −γdt (2.10)
ln
vfy + g/γ
v0y + g/γ
= −γtf ,
vfy = −g
γ +
v0y +
g
γ
e−γt. (2.11)
Although this direct integration is simpler than the method we invoke below, the method below
will come in useful for some slightly more difficult differential equations in the future. Thedifferential equation for vx is straight-forward to solve. Since it is first order there is one arbitraryconstant, A, and by inspection the solution is
vx = Ae−γt. (2.12)
The arbitrary constants for equations of motion are usually determined by the initial conditions,or more generally boundary conditions. By inspection A = v0x, the initial x component of thevelocity.
10
8/19/2019 Classical Mechanics Lectures
13/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
The differential equation for vy is a bit more complicated due to the presence of g. Differentialequations where all the terms are linearly proportional to a function, in this case vy, or to deriva-tives of the function, e.g., vy, dvy/dt, d
2vy/dt2 · · · , are called linear differential equations. If
there are terms proportional to v 2, as would happen if the drag force were proportional to thesquare of the velocity, the differential equation is not longer linear. Since this expression has onlyone derivative in v it is a first-order linear differential equation. If a term were added propor-
tional to d2
v/dt2
it would be a second-order differential equation. In this case we have a termcompletely independent of g, and the usual strategy is to first rewrite the equation with all thelinear terms on one side of the equal sign,
dvy
dt+ γvy = −g. (2.13)
Now, the solution to the equation can be broken into two parts. Since this is a first-order differen-tial equation we know that there will be one arbitrary constant. Physically, the arbitrary constantwill be determined by setting the initial velocity, though it could be determined by setting thevelocity at any given time. Like most differential equations, solutions are not "solved". Instead,one guesses at a form, then shows the guess is correct. For these types of equations, one firsttries to find a single solution, i.e. one with no arbitrary constants. This is called the particularsolution, yp(t), though it should really be called "a" particular solution because there are an in-finite number of such solutions. One then finds a solution to the "homogenous" equation, whichis the equation with zero on the right-hand side,
dvy,h
dt+ γvy,h = 0. (2.14)
The homogenous solution will have arbitrary constants. Since the sum of the two,
vy = vy,p + vy,h, (2.15)
is a solution of the total equation, one has now found a general form for all solutions, becausethe general solution is a solution of the differential equation, plus it has an arbitrary constantthat can be adjusted to fit any initial condition.
Returning to the example above, the homogenous solution is the same as that for vx, since therewas no gravitational acceleration in that case,
vy,h = Be−γt. (2.16)
In this case a particular solution is one with constant velocity,
vy,p = −g/γ. (2.17)Note that this is the terminal velocity of a particle falling from a great height. The general solu-tion is thus,
vy = Be−γt − g/γ, (2.18)
and one can find B from the initial velocity,
v0y = B − g/γ , B = v0y + g/γ. (2.19)
11
8/19/2019 Classical Mechanics Lectures
14/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
Plugging in the expression for B into Eq. (2.18) gives the y motion given the initial velocity,
vy = (v0y + g/γ )e−γt − g/γ. (2.20)
It is easy to see that this solution has vy = v0y when t = 0 and vy = −g/γ when t → ∞.One can also integrate the two equations to find the coordinates x and y as functions of t,
x =
t
0
dt v0x(t) =
v0x
γ
1 − e−γt , (2.21)
y =
t0
dt v0y(t) = −gt
γ +
v0y + g/γ
γ
1 − e−γt .
If the question was to find the position at a time t, we would be finished. However, the morecommon goal in a projectile equation problem is to find the range, i.e. the distance x at whichy returns to zero. For the case without a drag force this was much simpler. The solution forthe y coordinate would have been y = v0yt − gt2/2. One would solve for t to make y = 0,which would be t = 2v0y/g, then plug that value for t into x = v0xt to find x = 2v0xv0y/g =
v0 sin(2θ0)/g. One follows the same steps here, except that the expression for y(t) is morecomplicated. Searching for the time where y = 0, Eq. (2.21) gives
0 = −gtγ
+ v0y + g/γ
γ
1 − e−γt . (2.22)
This cannot be inverted into a simple expression t = · · · . Such expressions are known astranscendental equations, and are not the rare instance, but are the norm. In the days beforecomputers, one might plot the right-hand side of the equation graphically as a function of time,then find the point where it crosses zero.
Now, the most common way to solve for an equation of type Eq. (2.22) would be to applyNewton’s method numerically. This involves the following algorithm for finding solutions of some equation F (t) = 0.
1. First guess a value for the time, tguess.
2. Calculate F and its derivative, F (tguess) and F (tguess).
3. Unless you guessed perfectly, F = 0, and assuming that ∆F ≈ F ∆t, one would choose4. ∆t = −F (tguess)/F (tguess).
5. Now repeat step 1, but with tguess → tguess + ∆t.One breaks the loop once one finds F within sum acceptable tolerance of zero. A program to dothis might look like:
#include
#include
void CalcF(double t,double &F,double &Fprime);
12
8/19/2019 Classical Mechanics Lectures
15/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
using namespace std;
int main(){
int ntries=0;
double tguess,tolerance=1.0E-8,F,Fprime,delt;
printf("Enter tguess: ");
scanf("%lf",&tguess);CalcF(tguess,F,Fprime);
do{
delt=-Fprime/F;
if(fabs(delt)>0.5*fabs(tguess))
delt=0.5*tguess*delt/fabs(delt);
tguess=tguess+delt;
CalcF(tguess,F,Fprime);
ntries+=1;
}while(fabs(F)>tolerance && ntries
8/19/2019 Classical Mechanics Lectures
16/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
For a field in the z direction B = B ẑ, the force can only have x and y components,
F x = qBvy (2.24)
F y = −qBvx.The differential equations are
v̇x = ωcvy, ωc ≡ qB/m (2.25)v̇y = −ωcvx. (2.26)
One can solve the equations by taking time derivatives of either equation, then substituting intothe other equation,
v̈x = ωc v̇y = −ω2c vx. (2.27)The solution to these equations can be seen by inspection,
vx = A sin(ωct + φ), (2.28)
vy = A cos(ωct + φ).
One can integrate the equations to find the positions as a function of time,
x − x0 = x
x0
dx =
t0
dtv(t) (2.29)
= −A
ωccos(ωct + φ),
y − y0 = A
ωcsin(ωct + φ).
The trajectory is a circle centered at x0, y0 with amplitude A rotating in the clockwise direction.
The equations of motion for the z motion are
v̇z = 0, (2.30)
which leads toz − z0 = V zt. (2.31)
Added onto the circle, the motion is helical.
Note that the kinetic energy,
T = 1
2m(v2x + v
2y + v
2z) =
1
2m(ω2c A
2 + V 2z ), (2.32)
is constant. This is because the force is perpendicular to the velocity, so that in any differential
time element dt the work done on the particle F · dr = dt F · v = 0.One should think about the implications of a velocity dependent force. Suppose one had aconstant magnetic field in deep space. If a particle came through with velocity v0, it wouldundergo cyclotron motion with radius R = v0/ωc. However, if it were still its motion wouldremain fixed. Now, suppose an observer looked at the particle in one reference frame where theparticle was moving, then changed their velocity so that the particle’s velocity appeared to bezero. The motion would change from circular to fixed. Is this possible?
14
8/19/2019 Classical Mechanics Lectures
17/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
2.4 Conservation of Energy
Energy conservation is most convenient as a strategy for addressing particles where time doesnot appear. For example, a particle goes from position x0 with speed v0, to position xf – whatis its new speed? However, it can also be applied to problems where time does appear, such asin solving for the trajectory x(t), or equivalently t(x). This is the focus of this sub-section.
Energy is conserved in the case where the potential energy, U (r), depends only on position, andnot on time. The force is determined by U ,
F (r) = −∇U (r). (2.33)The net energy, E = U + T , is then conserved,
a = b (2.34)d
dt(T + U ) =
d
dt
m
2 (v2x + v
2y + v
2z) + U (r)
(2.35)
= mvx dvxdt
+ vydvy
dt
+ vzdvz
dt + ∂ xU dx
dt
+ ∂ yU dy
dt
+ ∂ zU dz
dt= vxF x + vyF y + vzF z − F xvx − F yvy − F zvz = 0. (2.36)
The same proof can be written more compactly with vector notation,
d
dt
m
2 v2 + U (r)
= mv · ̇v + ∇U (r) · ̇r (2.37)
= v · F − F · v = 0.
Inverting the expression for kinetic energy,
v =
2T /m =
2(E − U )/m, (2.38)one can solve for the one-dimensional trajectory x(t), by finding t(x),
t =
xx0
dx
v(x) =
xx0
dx 2(E − U (x))/m. (2.39)
Note this would be much more difficult in higher dimension, because you would have to deter-mine at what point x, y, z, the particles might reach in the trajectory, whereas in one dimensionyou can typically tell by simply seeing whether the kinetic energy is positive at the new position.
Example 2.2:Consider a simple harmonic oscillator potential, U (x) = (k/2)x2, with a particle emitted fromx = 0 with velocity v0. Solve for the trajectory t(x),
t =
x0
dx 2(E − kx2/2)/m (2.40)
=
m/k
x0
dx x2max − x2
, x2max = 2E/k.
15
8/19/2019 Classical Mechanics Lectures
18/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
Here E = mv20/2 and xmax is defined as the maximum displacement before the particle turnsaround. This integral is done by the substitution sin θ = x/xmax.
(k/m)1/2t = sin−1(x/xmax), (2.41)
x = xmax sin ωt, ω =
k/m.
2.5 Solving an integral numerically
As an example of an integral to solve numerically, consider the integral in Eq. (2.39). First,rewrite the integral as a sum,
t =N
n=1
∆x[2(E − U (xn)/m]−1/2, ∆x = x/N, xn = (n − 1/2)∆x. (2.42)
Note that for best accuracy the value of xn has been placed in the center of the nth interval. The
accuracy will improve for higher values of N . The pertinent part of a C++ program might looksomething like:
#include
#include
using namespace std;
double U(double x){
return "whatever function you like";
}
int main(){
int n,N=100;
double Deltax,x,x_n,t=0.0,E,mass=2.0;
double x0,xf,v0; // intial and final positions and initial velocity
printf("Enter initial and final positions: ");
scanf("%lf %lf",&x0,&xf);
printf("Enter initial velocity: ");
scanf("%lf",&v0);
E=0.5*mass*v0*v0+U(x0);
x=x0;
Deltax=(xf-x0)/N;for(x_n=x0+0.5*Deltax;x_n
8/19/2019 Classical Mechanics Lectures
19/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
2.6 Conservation of Momentum
Newton’s third law, "For every action there is an equal and opposite reaction", is more accuratelystated in the text,
If two bodies exert forces on each other, these forces are equal in magnitude and opposite indirection.
This means that for two bodies i and j, if the force on i due to j is called F ij , then
F ij = − F ji. (2.43)
Newton’s second law, F = ma, can be written for a particle i as
F i = j
F ij = miai, (2.44)
where F i (a single subscript) denotes the net force acting on i. Since the mass of i is fixed, one
can see that F i =
d
dtmivi =
j =i
F ij. (2.45)
Now, one can sum over all the particles and obtain
d
dt
i
mivi =ij
F i = j = 0. (2.46)
The last step made use of the fact that for every term ij, there is an equivalent term ji withopposite force. Since the momentum is defined as mv, for a system of particles
d
dt
i
mivi = 0, for isolated particles. (2.47)
By "isolated" one means that the only force acting on any particle i are those originating fromother particles in the sum.
Example 2.3:(Note that in the text, the rocket problem is discussed much later, along with collisions) Considera rocket of mass M 0 that ejects gas from the bottom of the rocket with a speed ve relative to therocket. The rocket’s mass reduces as the fuel is ejected by a rate α = dM/dt. Find the speed of the rocket as a function of time in terms of the initial mass and α. Neglect gravity.
Consider the rocket of mass M moving with velocity v. After a brief instant, the velocity of therocket is v + ∆v and the mass is M − ∆M . Momentum conservation gives
M v = (M − ∆M )(v + ∆v) + ∆M (v − ve)0 = −∆Mv + M ∆v + ∆M (v − ve),0 = M ∆v − ∆M ve.
17
8/19/2019 Classical Mechanics Lectures
20/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
In the second step we ignored the term ∆M ∆v because it is doubly small. The last equationgives
∆v = ve
M ∆M.
Integrating the expression with lower limits v0 = 0 and M 0, one finds
v = ve M
M 0dM
M
v = −ve ln(M/M 0)= −ve ln[(M 0 − αt)/M 0].
Since the total momentum of an isolated system is a constant, one can also quickly see that thecenter of mass of an isolated system is also constant. The center of mass is the average positionof a set of masses weighted by the mass,
x̄ =
i mixi
i mi. (2.48)
The rate of change of x̄ is
˙̄x = 1
M
i
mi ẋi = 1
M P x. (2.49)
Thus if the total momentum is constant the center of mass moves at a constant velocity, and if the total momentum is zero the center of mass is fixed.
2.7 Conservation of Angular Momentum
Consider a case where the force always points radially,
F (r) = F (r)r̂, (2.50)
where r̂ is a unit vector pointing outward from the origin. The angular momentum is defined as
L = r × p = mr × v. (2.51)
The rate of change of the angular momentum is
d L
dt= mv × v + mr × ̇v (2.52)= mv × v + r × F = 0.
The first term is zero because v is parallel to itself, and the second term is zero because F isparallel to r.
18
8/19/2019 Classical Mechanics Lectures
21/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
As an aside, one can see from the Levi-Civita symbol that the cross product of a vector with itself is zero.
V i = ( A × A)i = jk
ijkA jAk. (2.53)
For any term i, there are two contributions. For example, for i denoting the x direction, either j
denotes they
direction andk
denotes thez
direction, or vice versa, soV 1 = 123A2A3 + 132A3A2. (2.54)
This is zero by the antisymmetry of under permutations.
If the force is not radial, r × F = 0 in Eq. (2.52), and angular momentum is no longer conserved,
d L
dt= r × F ≡ τ , (2.55)
where τ is the torque.
For a system of isolated particles, one can write
ddt
i
Li =i = j
ri × F ij (2.56)
= 1
2
i = j
ri × F ij + r j × F ji
= 1
2
i = j
(ri − r j) × F ij = 0.
If the forces between the particles are radial, i.e. F ij || (ri − r j), then each term in the sum iszero and the net angular momentum is fixed. Otherwise, you could imagine an isolated system
that would start spinning spontaneously.
2.8 Exercises
1. Consider a bicyclist with air resistance proportional to v2 and rolling resistance propor-tional to v, so that
dv
dt= −Bv2 − Cv.
If the cyclist has initial velocity v0 and is coasting on a flat course, a) find her velocity as afunction of time, and b) find her position as a function of time.
2. For Eq. (2.22) show that in the limit where γ → 0 one finds t = 2v0y/g.3. The motion of a charged particle in an electromagnetic field can be obtained from the
Lorentz equation. If the electric field vector is E and the magnetic field is B, the forceon a particle of mass m that carries a charge q and has a velocity v
F = qE + qv × Bwhere we assume v
8/19/2019 Classical Mechanics Lectures
22/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
(a) If there is no electric field and if the particle enters the magnetic field in a directionperpendicular to the lines of magnetic flux, show that the trajectory is a circle withradius
r = mv
qB=
v
ωc,
where ωc ≡ qB/m is the cyclotron frequency.(b) Choose the z-axis to lie in the direction of B and let the plane containing E and B be
the yz−plane. ThusB = Bk̂, E = E yˆ j + E zk̂.
Show that the z component of the motion is given by
z(t) = z0 + ż0t + qE z
2mt2,
wherez(0) ≡ z0 and ż(0) ≡ ż0.
(c) Continue the calculation and obtain expressions for ẋ(t) and ẏ(t). Show that the timeaverages of these velocity components are
ẋ = E yB
, ẏ = 0.
(Show that the motion is periodic and then average over one complete period.)
(d) Integrate the velocity equations found in (c) and show (with the initial conditionsx(0) = −A/ωc, ẋ(0) = E y/B, y(0) = 0, ẏ(0) = A that
x(t) = −A
ωccos ωct +
E y
B
t, y(t) = A
ωcsin ωct.
These are the parametric equations of a trochoid. Sketch the projections of the tra- jectory on the xy-plane for the cases (i) A > |E y/B|, (ii) A < |E y/B|, and (iii)A = |E y/B|. (The last case yields a cycloid.)
4. A particle of mass m has velocity v = α/x, where x is its displacement. Find the forceF (x) responsible for the motion.
5. A particle is under the influence of a force F = −kx + kx3/α2, where k and α areconstants and k is positive. Determine U (x) and discuss the motion. What happens whenE = (1/4)kα2?
6. Using Eq. (2.39) find the position as a function of a time by numerical integration for the
case where a particle of mass m moves under the potential U (x) = U 0
x/L and foran initial velocity v0. Use the following values: m = 2.5kg, v0 = 75.0, L = 10m,U 0=15 J. Solve for time until the particle returns to the origin given that it started at theorigin moving with velocity v0. Make a graph of x vs t for the entire trajectory from yourcomputer output. Turn in the graph and a printout of your program.
20
8/19/2019 Classical Mechanics Lectures
23/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
7. Consider a rocket with initial mass M 0 at rest in deep space. It fires its engines which ejectmass with an exhaust speed ve relative to the rocket. The rocket loses mass at a constantrate α = dM/dt. Find the speed of the rocket as a function of time.
8. Imagine that a rocket can be built so that the best percentage of fuel to overall mass is 0.9.Explain the advantage of having stages.
9. Ted and his iceboat have a combined mass of 200 kg. Ted’s boat slides without friction onthe top of a frozen lake. Ted’s boat has a winch and he wishes to wind up a long heavyrope of mass 300 kg and length 100 m that is laid out in a straight line on the ice. Ted’s boatstarts at rest at one end of the rope, then brings the rope on board the ice boat at a constantrate of 0.25 m/s. After 400 seconds the rope is all aboard the iceboat.
(a) Before Ted turns on the winch, what is the position of the center of mass relative tothe boat?
(b) Immediately after Ted starts the winch, what is his speed?
(c) Immediately after the rope is entirely on the boat, what is Ted’s speed?
(d) Immediately after the rope is entirely on board, what is Ted’s displacement relative tohis original position?
(e) Immediately after the rope is entirely on board, where is the center of mass comparedto Ted’s original position?
(f) Find Ted’s velocity as a function of time.
10. Two disks are initially at rest, each of mass M , connected by a string between their centers.The disks slide on low-friction ice as the center of the string is pulled by a string with aconstant force F through a distance d. The disks collide and stick together, having moveda distance b horizontally. Determine the final speed of the disks just after they collide. Youmay want to use the proof from No. (12).
11. Santa Claus is skating on the magic ice near the north pole, which is frictionless. A masslessrope sticks out from the pole horizontally along a straight line. The rope’s original lengthis L0. Santa approaches the rope moving perpendicular to the direction of the rope andgrabs the end of the rope. The rope then winds around the thin pole until Santa is half theoriginal distance, L0/2, from the pole. If Santa’s original speed was v0, what is his newspeed?
21
8/19/2019 Classical Mechanics Lectures
24/74
PHY 321 Lecture Notes 2 NEWTON’S LAWS
12. Prove that the work on the center of mass during a small time interval δt, which is defined by
∆W cm ≡ F tot · δrcm,is equal to the change of the kinetic energy of the center of mass,
T cm ≡ 1
2M totv
2
cm, ∆T cm = M totvcm · δvcm.Use the following definitions:
F tot =
i
F i,
M tot =
i
mi,
rcm = 1
M tot
i
miri,
vcm = 1M tot
i
mivi.
22
8/19/2019 Classical Mechanics Lectures
25/74
PHY 321 Lecture Notes 3 OSCILLATIONS
3 Oscillations
3.1 Harmonic Oscillator
The harmonic oscillator seems to be omnipresent in physics. Although students think of this
as being related to springs, it appears in just about any problem where a mode is sitting nearits potential energy minimum. At that point, ∂ xV (x) = 0, and the first non-zero term (asidefrom a constant) in the potential energy is that of a harmonic oscillator. In a solid, sound modes(phonons) are built on a picture of coupled harmonic oscillators, and in relativistic field the-ory the fundamental interactions are also built on coupled oscillators positioned infinitesimallyclose to one another in space. The phenomena of a resonance of an oscillator driven at a fixedfrequency plays out repeatedly in atomic, nuclear and high-energy physics, when quantum me-chanically the evolution of a state oscillates according to e−iEt and exciting discrete quantumstates has very similar mathematics as exciting discrete states of an oscillator.
The potential energy for a single particle as a function of its position x can be written as a Taylorexpansion about some point x0
V (x) = V (x0) + (x − x0) ∂ xV (x)|x0 + 1
2(x − x0)2 ∂ 2xV (x)
x0
+ 1
3! ∂ 3xV (x)
x0
+ · · · (3.1)
If the position x0 is at the minimum of the resonance, the first two terms of the potential are
V (x) ≈ V (x0) + 1
2(x − x0)2 ∂ 2xV (x)
x0
, (3.2)
= V (x0) + 1
2k(x − x0)2, k ≡ ∂ 2xV (x)
x0
,
F = −∂ xV (x) = −k(x − x0).
Put into Newton’s 2nd law (assuming x0 = 0),
mẍ = −kx, (3.3)x = A cos(ω0t − φ), ω0 =
k/m. (3.4)
Here A and φ are arbitrary. Equivalently, one could have written this as A cos(ω0t)+B sin(ω0t),or as the real part of Aeiω0t. In the last case A could be an arbitrary complex constant. Thus,there are 2 arbitrary constants, or the real and imaginary part of one complex constant. This isthe expectation for a second order differential equation, and also agrees with the physical expec-tation that if you know a particle’s initial velocity and position you should be able to define its
future motion.One can also calculate the velocity and the kinetic energy as a function of time,
ẋ = −ω0A sin(ω0t − φ), (3.5)T =
1
2mẋ2 =
mω20A2
2 sin2(ω0t − φ),
= k
2A2 sin2(ω0t − φ).
23
8/19/2019 Classical Mechanics Lectures
26/74
PHY 321 Lecture Notes 3 OSCILLATIONS
The total energy is then
E = T + U = 1
2mẋ2 +
1
2kx2 =
1
2kA2. (3.6)
The total energy then goes as the square of the amplitude and the frequency is independent of the amplitude.
Example 3.1:A pendulum is an example of a harmonic oscillator. By expanding the kinetic and potentialenergies for small angles find the frequency for a pendulum of length L with all the mass mcentered at the end by writing the eq.s of motion in the form of a harmonic oscillator.
Solution: The potential energy and kinetic energies are (for x being the displacement)
U = mgL(1 − cos θ) ≈ mgL x2
2L2,
T = 1
2mL2 θ̇2
≈ m
2 ẋ2.
For small x Newton’s 2nd law becomes
mẍ = −mgL
x,
and the spring constant would appear to be k = mg/L, which makes the frequency equal to
ω0 =
g/L. Note that the frequency is independent of the mass.
3.2 Damped Oscillators
In this chapter we consider only the case where the damping force is proportional to the ve-locity. This is counter to dragging friction, where the force is proportional in strength to thenormal force, and is also inconsistent with wind resistance. Rolling resistance does seem to beproportional to the velocity. However, the main motivation for considering damping forces pro-portional to the velocity is that the math is more friendly. This is because the differential equationis linear, i.e. each term is of order x, ẋ, ẍ · · · , or even terms with no mention of x, and there areno terms such as x2 or xẍ. The equations of motion for a spring with damping force −bẋ are
mẍ + bẋ + kx = 0. (3.7)
Just to make the solution a bit less messy, we rewrite this equation as
ẍ + 2βẋ + ω20x = 0, β ≡ b/2m, ω0 ≡
k/m. (3.8)
Both β and ω have dimensions of inverse time. To find solutions (see appendix C in the text)you must make an educated guess at the form of the solution. To do this, first realize that the
24
8/19/2019 Classical Mechanics Lectures
27/74
PHY 321 Lecture Notes 3 OSCILLATIONS
solution will need an arbitrary normalization A because since the equation is linear. Secondly,realize that if the form is
x = Aert (3.9)
that each derivative simply brings out an extra power of r. This means that the Aert factors outand one can simply solve for an equation for r. Plugging this form into Eq. (3.8),
r2
+ 2βr + ω2
0 = 0. (3.10)Since this is a quadratic equation there will be two solutions,
r = −β ±
β2 − ω20. (3.11)
We refer to the two solutions as r1 and r2 corresponding to the + and − roots. As expected,there should be two arbitrary constants involved in the solution,
x = A1er1t + A2e
r2t, (3.12)
where the coefficients A1 and A2 are determined by initial conitions.
The roots listed above can be imaginary if the damping is small and ω0 > β. There are threecases:
1. Underdamped: β < ω0
x = A1e−βteiω
t + A2e−βte−iω
t, ω ≡
ω20 − β2 (3.13)= (A1 + A2)e
−βt cos ωt + i(A1 − A2)e−βt sin ωt. (3.14)Here we have made use of the identity eiω
t = cos ωt + i sin ωt. Since the constantsare arbitrary, and since the real and imaginary parts are both solutions individually, I can
simply consider the real part of the solution alone:x = B1e
−βt cos ωt + B2e−βt sin ωt. (3.15)
2. Critical dampling: β = ω0In this case the two terms involving r1 and r2 are identical because ω = 0. Since weneed to arbitrary constants, there needs to be another solution. This is found by simplyguessing, or by taking the limit of ω → 0 from the underdamped solution. The solutionis then
x = Ae−βt + Bte−βt. (3.16)
The critically damped solution is interesting because the solution approaches zero quickly,
but does not oscillate. For a problem with zero initial velocity, the solution never crosseszero. This is a good choice for designing shock absorbers or swinging doors.
3. Overdamped: β > ω0
x = A1e−(β+
√ β2−ω20)t + A2e
−(β−√
β2−ω20)t (3.17)
This solution will also never pass the origin more than once, and then only if the initialvelocity is strong and initially toward zero.
25
8/19/2019 Classical Mechanics Lectures
28/74
PHY 321 Lecture Notes 3 OSCILLATIONS
Example 3.2:Given b, m and ω0, find x(t) for a particle whose initial position is x = 0 and has initial velocityv0 (assuming an underdamped solution).
Solution: The solution is of the form,
x = e−βt
[A1 cos(ω
t) + A2 sin ω
t] ,ẋ = −βx + ωe−βt [−A1 sin ωt + A2 cos ωt] .
ω ≡
ω20 − β2, β ≡ b/2m.
From the initial conditions, A1 = 0 because x(0) = 0 and ωA2 = v0. So
x = v0
ωe−βt sin ωt.
3.3 Sinusoidally Driven Oscillators
Here, we consider the forceF = −kx − bẋ + F 0 cos ωt, (3.18)
which leads to the differential equation
ẍ + 2βẋ + ω20x = (F 0/m)cos ωt. (3.19)
Consider a single solution with no arbitrary constants, which we will call a particular solution,
xp(t). Now consider solutions to the same equation without the driving term, which includetwo arbitrary constants. These are called either homogenous solutions or complementary solutions,and were given in the previous section, e.g. Eq. (3.15) for the underdamped case. The homoge-nous solution already incorporates the two arbitrary constants, so any sum of a homogenoussolution and a particular solution will represent the general solution of the equation. To find theparticular solution, one first guesses at the correct form,
xp(t) = D cos(ωt − δ), (3.20)
and rewrite the differential equation as
D−ω2 cos(ωt − δ) − 2βω sin(ωt − δ) + ω20 cos(ωt − δ) = C cos(ωt), C ≡ F 0/m.
(3.21)One can now use angle addition formulas to get
D
(−ω2 cos δ + 2βω sin δ + ω20 cos δ) cos(ωt) (3.22)+(−ω2 sin δ − 2βω cos δ + ω20 sin δ) sin(ωt)
= C cos(ωt).
26
8/19/2019 Classical Mechanics Lectures
29/74
PHY 321 Lecture Notes 3 OSCILLATIONS
Both the cos and sin terms need to equate if the expression is to hold at all times. Thus, this becomes two equations
D−ω2 cos δ + 2βω sin δ + ω20 cos δ) = C (3.23)
−ω2 sin δ − 2βω cos δ + ω20 sin δ = 0.
The lower expression leads to
tan δ = 2βω
ω20 − ω2, (3.24)
sin δ = 2βω
(ω20 − ω2)2 + 4ω2β2,
cos δ = (ω20 − ω2) (ω20 − ω2)2 + 4ω2β2
(3.25)
Inserting the expressions for cos δ and sin δ into the expression for D,
D = C (ω20 − ω2)2 + 4ω2β2
. (3.26)
For a given initial condition, e.g. initial displacement and velocity, one must add the homoge-nous solution then solve for the two arbitrary constants. However, because the homogenoussolutions decay with time as e−βt, the particular solution is all that remains at large times, andis therefore the steady state solution. Since the arbitrary constants are all in the homogenoussolution, all memory of the initial conditions are lost at large times, t >> 1/β.
The amplitude of the motion, D, is linearly proportional to the driving force (C = F 0/m), butalso depends on the driving frequency ω. For small β the maximum will occur at ω = ω0. This
is referred to as a resonance. In the limit β → 0 the amplitude at resonance approaches infinity.
3.4 Alternative Derivation for Driven Oscillators
Here we will assume a complex form for the driving force and for the particular solution
F (t) = F 0eiωt, xp = De
iωt, (3.27)
where D is a complex constant. Note that the differential equation for xp has both a real andimaginary solution, we will effectively solve two problems at once. The real part of the solution
will correspond to a driving force of (F 0/m)cos ωt, and the imaginary part will correspond toa driving force of (F 0/m)sin ωt. After finding the solution, we will take the real part of xp toget the solution we are after in this case.
From Eq. (3.19) one inserts the form for xp above to get
D−ω2 + 2iβω + ω20 eiωt = C eiωt, (3.28)
D = C
(ω20 − ω2) + 2iβω.
27
8/19/2019 Classical Mechanics Lectures
30/74
PHY 321 Lecture Notes 3 OSCILLATIONS
The norm and phase for D = |D|e−iδ can be read by inspection,
|D| = 1 (ω20 − ω2)2 + 4β2ω2
, tan δ = 2βω
ω20 − ω2. (3.29)
This is the same expression for δ as before. Once then finds xp(t),
xp(t) = Ceiωt−iδ
(ω20 − ω2)2 + 4β2ω2(3.30)
= C cos(ωt − δ)
(ω20 − ω2)2 + 4β2ω2. (3.31)
This is the same answer as before. If one wished to solve for the case where F (t)/m = C sin ωt,the imaginary part of the solution would work
xp(t) = Ceiωt−iδ
(ω20 − ω2)2 + 4β2ω2
(3.32)
= C sin(ωt − δ)
(ω20 − ω2)2 + 4β2ω2. (3.33)
Example 3.3:Consider the damped and driven harmonic oscillator worked out above. Given F 0, m , β andω0, solve for the complete solution x(t) for the case where F = F 0 sin ωt with initial conditionsx(t = 0) = 0 and v(t = 0) = 0. Assume the underdamped case.
Solution: The entire solution including the arbitrary constants includes both the homogenousand particular solutions,
x(t) = F 0
m
sin(ωt − δ) (ω20 − ω2)2 + 4β2ω2
+ A cos ωte−βt + B sin ωte−βt.
The quantities δ and ω are given earlier in the section, ω =
ω20 − β2, δ = tan−1(2βω/(ω20 −ω2). Here, solving the problem means finding the arbitrary constants A and B . Satisfying theinitial conditions for the initial position and velocity:
x(t = 0) = 0 = −η sin δ + A,v(t = 0) = 0 = ωη cos δ − βA + ωB,
η ≡ F 0
m
1 (ω20 − ω2)2 + 4β2ω2 .
The problem is now reduced to 2 equations and 2 unknowns, A and B. The solution is
A = η sin δ, B = −ωη cos δ + βη sin δ
ω . (3.34)
28
8/19/2019 Classical Mechanics Lectures
31/74
PHY 321 Lecture Notes 3 OSCILLATIONS
ω
F W H M
Figure 3.1: The maximum amplitude squared of the steady-state motion of a sinusoidally driven har-monic oscillator is shown as a function of the driving frequency ω. This peaks near the fundamentalfrequency ω0 and the full-width-half-maximum, F W H M , is given by the damping, 2β.
3.5 Resonance Widths; the Q factor
From the previous two sections, the particular solution for a driving force F = F 0 cos ωt is
xp(t) = 1
(ω20 − ω2)2 + 4ω2β2F 0
mcos(ωt − δ), (3.35)
δ = tan−1
2βω
ω20 − ω2
.
If one fixes the driving frequency ω and adjusts the fundamental frequency ω0 =
k/m, themaximum amplitude occurs when ω0 = ω because that is when the term from the denominator(ω20 − ω2)2 + 4ω2β2 is at a minimum. This is akin to dialing into a radio station. However, if one fixes ω0 and adjusts the driving frequency one minimize with respect to ω, e.g. set
d
dω
(ω20 − ω2)2 + 4ω2β2
= 0, (3.36)
and one finds that the maximum amplitude occurs when ω =
ω20 − 2β2. If β is small relativeto ω0, one can simply state that the maximum amplitude is
xmax ≈ F 0
2mβω0. (3.37)
Figure 3.1 displays the maximum amplitude squared as a function of ω, and one can see the peakat ω
≈ ω
0. The squared amplitude is usually more the quantity of interest than the amplitude
because the power being absorbed by the oscillator is proportional to the square of the ampli-tude, not the amplitude. The width of the peak is quantified by the full-width-half-maximum,sometimes called F W H M and can be found by finding the frequency difference, ω − ω, suchthat the maximum response falls by a factor of two,
4ω2β2
(ω20 − ω2)2 + 4ω2β2 =
1
2. (3.38)
29
8/19/2019 Classical Mechanics Lectures
32/74
PHY 321 Lecture Notes 3 OSCILLATIONS
For small damping this occurs when ω = ω0 ± β, so the F W H M ≈ 2β. For the purposes of tuning to a specific frequency, one wants the width to be as small as possible. The ratio of ω0 toF W H M is known as the quality factor, or Q factor,
Q ≡ ω02β
. (3.39)
3.6 Principal of Superposition and Periodic Forces
If one has several driving forces, F (t) =
n F n(t), one can find the particular solution to eachone, xpn(t), then the particular solution for the entire driving force is
xp(t) =n
xpn(t). (3.40)
This is known as the principal of superposition and in only applies when the homogenous equa-tion is linear. If there were an anharmonic term such as x3 in the homogenous equation, then
when one summed various solutions, x = (
n xn)2
, one would get cross terms. Superpositionis especially useful when F (t) can be written as a sum of sinusoidal terms, because the solutionsfor each sinusoidal term is analytic, and are given in the previous two subsections.
Driving forces are often periodic, even when they are not sinusoidal. Periodicity implies that forsome time τ
F (t + τ ) = F (t). (3.41)
One example of a non-sinusoidal periodic force is a square wave. Many components in electriccircuits are non-linear, e.g. diodes, which makes many wave forms non-sinusoidal even whenthey circuits are being driven by purely sinusoidal sources.
For the sinusoidal example studied in the previous subsections the period is τ = 2π/ω. How-ever, higher harmonics can also satisfy the periodicity requirement. In general, any force thatsatisfies the periodicity requirement can be expressed as a sum over harmonics,
F (t) = f 0
2 +n>0
f n cos(2nπt/τ ) + gn sin(2nπt/τ ). (3.42)
From the previous subsection, one can write down the answer for xpn(t), by substituting f n/mor gn/m for C into Eq.s (3.30) o r (3.32) respectively. By writing each factor 2nπt/τ as nωt, withω ≡ 2π/τ ,
F (t) = f 02
+n>0
f n cos(nωt) + gn sin(nωt). (3.43)
30
8/19/2019 Classical Mechanics Lectures
33/74
PHY 321 Lecture Notes 3 OSCILLATIONS
The solutions for x(t) then come from replacing ω with nω for each term,
xp(t) = f 0
2k+n>0
αn cos(nωt − δn) + βn sin(nωt − δn), (3.44)
αn = f n/m
((nω)2 − ω20) + 4β
2n2ω2,
βn = gn/m
((nω)2 − ω20) + 4β2n2ω2,
δn = tan−1
2βnω
ω20 − n2ω2
. (3.45)
The coefficients f n and gn can be extracted from the function F (t) by
f n = 2
τ
τ/2−τ/2
dt F (t) cos(2nπt/τ ), (3.46)
gn = 2τ
τ/2−τ/2
dt F (t) sin(2nπt/τ ).
To check the consistency of these expressions, one can insert the expansion of F (t) in Eq. (3.43)into the expression for the coefficients in Eq. (3.46) and see whether
f n =? 2
τ
τ/2−τ/2
dt
f 0
2 +m>0
f m cos(mωt) + gn sin(mωt)
cos(nωt). (3.47)
Immediately, one can throw away all the terms with gm because they convolute an even and an
odd function. The term with f 0/2 disappears because cos(nωt) is equally positive and negativeover the interval and will integrate to zero. For all the terms f m cos(mωt) appearing in the sum,one can use angle addition formulas to see that cos(mωt) cos(nωt) = (1/2)(cos[(m+n)ωt]+cos[(m − n)ωt]. This will integrate to zero unless m = n. In that case the m − n term, whichgives cos(mωt) cos(nωt) = 1/2, and
f n = 2
τ
τ/2−τ/2
dt f n/2 (3.48)
= f n .
The same method can be used to check for the consistency of gn.
Example 3.4:Consider the driving force:
F (t) = At/τ, − τ /2 < t < τ/2, otherwise determined by periodicity (3.49)
Find the Fourier coefficients f n and gn for all n using Eq. (3.46).
31
8/19/2019 Classical Mechanics Lectures
34/74
PHY 321 Lecture Notes 3 OSCILLATIONS
1 1 1
Figure 3.2: The function periodic function F (t) = t/τ, |t| < τ /2 (black) is compared to the Fourierexpansion described in Eq. (3.50) with 10 terms (red) or 100 terms (green).
Solution: Only the odd coefficients enter by symmetry and f n = 0. One can find gn integrating by parts.
gn = 2
τ
τ/2−τ/2
dt sin(nωt)At
τ (3.50)
u = t, dv = sin(nωt)dt, v = − cos(nωt)/(nω),
gn = −2A
nωτ 2
τ/2−τ/2
dt cos(nωt) + 2A−t cos(nωt)
nωτ 2
τ/2
−τ/2
.
The first term is zero because cos(nωt) will be equally positive and negative over the interval.Using the fact that ωτ = 2π,
gn = − 2A
2nπcos(nωτ/2) (3.51)
= − Anπ
cos(nπ)
= A
nπ(−1)n+1.
The true function is compared to the expansion in Fig. 3.2.
3.7 Response to arbitrary, but temporary, force
Consider a particle at rest in the bottom of an underdamped harmonic oscillator, that then feelsa sudden impulse I = F ∆t at t = 0. This increases the velocity immediately by an amount
32
8/19/2019 Classical Mechanics Lectures
35/74
PHY 321 Lecture Notes 3 OSCILLATIONS
v0 = I/m while not changing the position. One can then solve the trajectory by solving Eq.(3.15) with initial conditions v0 = F ∆t/m and x0 = 0. This gives
x(t) = I
mωe−βt sin ωt, t > 0 (3.52)
For an impulse that occurs at time ti the trajectory would be
x(t) = I imω
e−β(t−ti) sin[ω(t − ti)], t > ti (3.53)If there were several impulses linear superposition tells us that we can sum over each contribu-tion,
x(t) =
i
I i
mωe−β(t−ti) sin[ω(t − ti)]Θ(t − ti), (3.54)
where Θ(x) = 0 for x 0.
Now one can consider a series of impulses at times separated by ∆t, where each impulse isgiven by F i∆t. The solution now becomes
x(t) =
i
F i∆t
mω e−β(t−ti) sin[ω(t − ti)]Θ(t − ti) (3.55)
=
t−∞
dt F (t)
mω e−β(t−t
) sin[ω(t − t)].
The quantity e−β(t−t) sin[ω(t − t)]/mω is called a Green’s function, G(t − t) as it describes
the response at t due to a force applied at an earlier time t. When performing the integral in Eq.(3.55) one can use angle addition formulas to factor out the part with the t dependence in theintegrand if, and only if, the time t is beyond the time at which the potential is still interacting,
x(t) = 1
mωe−βt [I c sin(ω
t)
−I s cos(ω
t)] , (3.56)
I c(t) ≡ t−∞
dt F (t)eβt
cos(ωt),
I s(t) ≡ t−∞
dt F (t)eβt
sin(ωt).
If the time t is beyond any time at which the force acts, F (t > t) = 0, the coefficients becomeconstants.
Example 3.5:
3.8 Exercises
1. A floating body of uniform cross-sectional area A and of mass density ρ and at equilibriumdisplaces a volume V . Show that the period of small oscillations about the equilibriumposition is given by
τ = 2π
V/gA
33
8/19/2019 Classical Mechanics Lectures
36/74
PHY 321 Lecture Notes 3 OSCILLATIONS
2. As a Taylor expansion in powers of t, express
(a) sin(ωt)
(b) cos(ωt)
(c) eiωt
3. Show that the critically damped solution, Eq. (3.16), is indeed the solution to the differen-tial equation.
4. Consider an over-damped harmonic oscillator with a mass of m = 2 kg, a damping factorb = 20 Ns/m, and a spring constant k = 32 N/m. If the initial position is x = 0.125m, and if the initial velocity is −2.0 m/s, find and graph the motion as a function of time.Solve for the time at which the mass crosses the origin.
5. Consider a particle of mass m moving in a one-dimensional potential,
V (x) = −k x2
2 + α
x4
4 .
(a) What is the angular frequency for small vibrations about the minimum of the poten-tial? What is the effective spring constant?
(b) If you add a small force F = F 0 cos(ωt − φ), and if the particle is initially at theminimum with zero initial velocity, find its position as a function of time.
(c) If there is a small drag force −bv, repeat (b).6. Consider the periodic force, F (t + τ ) = F (t),
F (t) =
−A, −τ /2 < t
8/19/2019 Classical Mechanics Lectures
37/74
PHY 321 Lecture Notes 3 OSCILLATIONS
(c) Using the definition of Fourier coefficients in Eq.s (3.43) and (3.46), show that
δ(t − t0) = −1
τ +
2
τ
∞n=0
cos(ωn(t − t0)), ωn = 2nπ/τ.
8. Consider the complex function in the interval −τ /2 < t < τ/2,
f (t) = −1τ
+ 2
τ
∞n=0
einω(t−t0), ω = 2π/τ.
(a) Using the fact that if one integrates over the interval, −τ /2 < t < τ /2, that dteinωt =0 for n = 0, show that
dtf (t) = 1.
(b) Using the fact that
n xn = 1/(1 − x), show that
f (t) = −1
τ +
2/τ
1 − eiω(t−t0) .(c) From the expression in (b), show that the real part of f (t) obeys
f (t) = 0, for t = t0This shows that f is a delta function and validates the result of the previous prob-lem.
9. A particle of mass m in an undamped harmonic oscillator with angular frequency ω0 is atrest in the bottom of the well, when it experiences a force
F (t) =
0, t τ
Find x(t) for t > τ .
10. Consider a particle of mass m in a harmonic oscillator with angular frequency ω0 and nodamping. It experiences an external force,
F (t) = f 0Θ(t)e−γt.
A “Theta” function is a step function, and is zero for negative arguments and unity forpositive arguments.
(a) Find a particular solution, xp(t), assuming it is proportional to e−γt .
(b) For a particle initially at rest at the origin at t = 0, find x(t) by adding in the homoge-nous solutions and matching the BC, determine the arbitrary constants.
(c) Use Eq. (3.55) or Eq. (3.56) to find x(t). Check that you get the same result as (b).
35
8/19/2019 Classical Mechanics Lectures
38/74
PHY 321 Lecture Notes 4 GRAVITY AND CENTRAL FORCES
4 Gravity and Central Forces
4.1 Gravity
The gravitational potential energy and forces involving two masses a and b are
U ab = −Gmamb
|ra − rb|, (4.1)
F ba = − Gmamb
|ra − rb|2r̂ab,
r̂ab = rb − ra|ra − rb|
.
Here G = 6.67 × 10−11 Nm2/kg2, and F ba is the force on b due to a. By construction, the forceon b due to a and the force on a due to b are equal and opposite. The net potential energy for a
large number of masses would be
U =a
8/19/2019 Classical Mechanics Lectures
39/74
PHY 321 Lecture Notes 4 GRAVITY AND CENTRAL FORCES
Solution: Take the ratio of Eq. (4.3) for two radii, R and r < R ,
4πr2g(r)
4πR 2g(R ) =
4πGM inside r
4πGM inside R
= r3
R 3
g(r) = g(R ) rR
.
The potential energy per mass is similar conceptually to the voltage, or electric potential energyper charge that was studied in electromagnetism. If V ≡ U/m, g = −∇V .
4.2 Tidal ForcesConsider a spherical object of radius r a distance D from another body of mass M . The magni-tude of the force due to M on an object of mass δm on surface of the planet can be calculated byperforming a Taylor expansion about the center of the spherical object.
F = −GMδmD2
+ 2GMδm
D3 ∆D + · · · (4.4)
If the z direction points toward the large object, ∆D can be referred to as z. In the acceleratingframe of an observer at the center of the object,
δm d2z
dt2 = F − δma + other forces acting on δm, (4.5)
where a is the acceleration of the observer. Since δma equals the gravitational force on δm if it were located at the object’s center, one can write
md2z
dt2 = 2
GMδm
D3 z + other forces acting on δm. (4.6)
Here the other forces could represent the forces acting on δm from the spherical object such asthe gravitational force of the contact force with the surface. If θ is the angle w.r.t. the z axis, the
effective force acting on δm is
F eff = 2GMδm
D3 r cos θ + other forces acting on δm. (4.7)
This first force is the "tidal" force. It pulls objects outward from the center of the object. If theobject were covered with water, it would distort the objects shape so that the shape would beelliptical, stretched out along the axis pointing toward the large mass M . The force is alwaysalong (either parallel or antiparallel to) the ẑ direction.
37
8/19/2019 Classical Mechanics Lectures
40/74
PHY 321 Lecture Notes 4 GRAVITY AND CENTRAL FORCES
Example 4.2:Consider the Earth to be a sphere of radius R covered with water, with the gravitational ac-celeration at the surface noted by g. Now assume that a distant body provides an additionalconstant gravitational acceleration a pointed along the z axis. Find the distortion of the radiusas a function of θ. Ignore planetary rotation and assume a
8/19/2019 Classical Mechanics Lectures
41/74
PHY 321 Lecture Notes 4 GRAVITY AND CENTRAL FORCES
The Moon more strongly affects tides than the Sun.
4.3 Deriving elliptical orbits
Kepler’s laws state that a gravitational orbit should be an ellipse with the source of the gravi-tational field at one focus. Deriving this is surprisingly messy. To do this, we first derive theequations of motion in terms of r and θ rather than in terms of x and y. The overall strategy isto
1. Find equations of motion for r and t with no angle (θ) mentioned, i.e. d2r/dt2 = · · · .2. Use angular momentum conservation to find an expression for θ̇ in terms of r.
3. Use the chain rule to convert the equations of motions for r, an expression involving r, ṙand r̈, to one involving r, dr/dθ and d2r/dθ2. This is made more complicated becausethe expressions will also involve a substitution u = 1/r so that one finds an expression in
terms of u and θ.4. Once u(θ) is found, you need to show that this can be converted to the familiar form for
an ellipse.
The equations of motion give
d
dtr2 =
d
dt(x2 + y2) = 2xẋ + 2yẏ, (4.8)
ṙ = x
rẋ +
y
rẏ,
r̈ =
x
r ẍ +
y
r ÿ +
ẋ2 + ẏ2
r − ṙ2
r
= F x cos θ + F y sin θ
m+
ṙ2 + r2 θ̇2
r− ṙ
2
r
= F
m+
r2 θ̇2
r
= F
m+
L2
m2r3.
This derivation used the fact that the force was radial, F = F r = F x cos θ + F y sin θ, and that
angular momentum is L = mrvθ = mr2 θ̇. The term L2/mr3 = mv2/r behaves appears
like an additional force. Sometimes this is referred to as a centrifugal force, but it is not a force, but merely the consequence of considering the motion in a rotating (and therefore accelerating)frame.
Now, we switch to the particular case of an attractive inverse square force, F = −α/r2, andshow that the trajectory, r(θ), is an ellipse. To do this we transform derivatives w.r.t. timeto derivatives w.r.t. θ using the chain rule combined with angular momentum conservation,
39
8/19/2019 Classical Mechanics Lectures
42/74
PHY 321 Lecture Notes 4 GRAVITY AND CENTRAL FORCES
θ̇ = L/mr2.
ṙ = dr
dθθ̇ =
dr
dθ
L
mr2, (4.9)
r̈ = d2r
dθ2
L
mr2
2− 2
r
dr
dθ
2 Lmr2
2
Equating the two expressions for r̈ in Eq. (4.8) and Eq. (4.9) provides a differential equation,
d2r
dθ2
L
mr2
2− 2
r
dr
dθ
2 Lmr2
2=
F
m+
L2
m2r3, (4.10)
that when solved yields the trajectory. Up to this point the expressions work for any radial force,not just forces that fall as 1/r2.
The trick to simplifying this differential equation for the inverse square problems is to make asubstitution, u ≡ 1/r, and rewrite the differential equation for u(θ).
r = 1/u, (4.11)dr
dθ=
−1u2
du
dθ,
d2r
dθ2 =
2
u3
du
dθ
2− 1
u2d2u
dθ2.
Plugging these expressions into Eq. (4.10) gives an expression in terms of u, du/dθ, and d2u/dθ2.After some tedious algebra,
d2u
dθ2 = −u − F m
L2u2. (4.12)
For the attractive inverse square law force, F = −αu2
,
d2u
dθ2 = −u + mα
L2 . (4.13)
The solution has two arbitrary constants, A and θ0,
u = mα
L2 + A cos(θ − θ0), (4.14)
r = 1
(mα/L2) + A cos(θ − θ0).
The radius will be at a minimum when θ = θ0 and at a maximum when θ = θ0 + π . Theconstant A is related to the eccentricity of the orbit. When A = 0 the radius is a constantr = L2/mα. If one solved the expression mv2/r = −α/r2 for a circular orbit, using thesubstitution v = L/mr, one would reproduce the expression above with A = 0.
The form describing the elliptical trajectory in Eq. (4.14) can be identified as an ellipse with onefocus being the center of the ellipse by considering the definition of an ellipse as being the points
40
8/19/2019 Classical Mechanics Lectures
43/74
PHY 321 Lecture Notes 4 GRAVITY AND CENTRAL FORCES
such that the sum of the two distances between the two foci are a constant. Making that distance2D, the distance between the two foci as 2a and putting one focus at the origin,
2D = r +
(r cos θ − 2a)2 + r2 sin2 θ, (4.15)4D2 + r2 − 4Dr = r2 + 4a2 − 4ar cos θ,
r = D2
−a2
D + a cos θ = 1
D/(D2 − a2) − a cos θ/(D2 − a2) .
By inspection, this is the same form as Eq. (4.14).
4.4 Effective or Centrifugal Potential
The total energy of a particle is
E = U (r) + 1
2mv2θ +
1
2mṙ2 (4.16)
= U (r) + 12
mr2 θ̇2 + 12
mṙ2
= U (r) + L2
2mr2 +
1
2mṙ2.
The second term then contributes to the energy like an additional repulsive potential. The termis sometimes referred to as the "centrifugal" potential or the "effective" potential, even though itis actually a potential energy. Note that if one treats the effective potential like a real potential,one would expect to be able to generate an effective force,
F eff = − d
dr
L2
2mr2 =
L2
mr3 = m
v2⊥r = mω
2
r, (4.17)
which is indeed the "centrifugal" force from Eq. (4.8).
Example 4.5:Consider a particle of mass m in a 2-dimensional harmonic oscillator with potential
U = 1
2kr2.
If the orbit has angular momentum L, find
a) the radius and angular velocity of the circular orbit b) the angular frequency of small radial perturbations
Solution:a) Consider the effective potential. The radius of a circular orbit is at the minimum of the poten-
41
8/19/2019 Classical Mechanics Lectures
44/74
PHY 321 Lecture Notes 4 GRAVITY AND CENTRAL FORCES
tial (where the effective force is zero).
U eff = 1
2kr2 +
L2
2mr2,
0 = krmin − L2
mr3min,
rmin =
L2
mk
1/4
,
θ̇ = L
mr2min=
k/m.
b) Now consider small vibrations about rmin. The effective spring constant is the curvature of the effective potential.
keff = d2
dr2U (r)
r=rmin= k +
3L2
mr4min= 4k,
ω =
keff /m = 2
k/m = 2 θ̇.
Unlike the inverse-square force, the harmonic oscillator has two minima and two maxima. Theorbits are elliptical, but the center of the ellipse coincides with r = 0.
The solution is also simple to write down exactly in Cartesian coordinates. The x and y equa-tions of motion separate,
ẍ = −kx,ÿ =
−ky.
So the general solution can be expressed as
x = A cos ω0t + B sin ω0t,
y = C cos ω0t + D sin ω0t.
With some work using double angle formulas, one can calculate
r2 = x2 + y2 = α + β cos2ω0t + γ sin2ω0t,
α = A2 + B2 + C 2 + D2
2 , β =
A2 − B2 + C 2 − D22
, γ = AB + CD,
r2 = α + (β2 + γ 2)1/2 cos(2ω0t − δ), δ = arctan(γ/β),
and see that the radius oscillates with frequency 2ω0. The factor of two comes because theoscillation x = A cos ω0t has to maxima for x2, one at t = 0 and one a half period later.
42
8/19/2019 Classical Mechanics Lectures
45/74
PHY 321 Lecture Notes 4 GRAVITY AND CENTRAL FORCES
r
U ( r ) = − α
1
/ r
U ( r ) = − α
2
/ r
3
V
e f f
( r ) = U ( r ) + L
2
/ 2 m r
2
Figure 4.1: The effective potential is sketched for two cases, a 1/r attractive potential and a 1/r3 attractivepotential. The 1/r case has a stable minimum, whereas the circular orbit in the 1/r3 case is unstable.
4.5 Stability of Orbits
The effective force can be extracted from the effective potential, U eff . Beginning from the equa-tions of motion, Eq. (4.8), for r,
mr̈ = F + L2
mr3 (4.18)
= F eff
= −∂ rU eff ,F eff = −∂ r
U (r) − (L2/2mr2) .
For a circular orbit, the radius must be fixed as a function of time, so one must be at a maximum
or a minimum of the effective potential. However, if one is at a maximum of the effective poten-tial the radius will be unstable. For the attractive Coulomb force the effective potential will bedominated by the −α/r term for large r since the centrifugal part falls off more quickly, ∼ 1/r2.At low r